June 23, 2022

Carson v. Makin and the Dwindling Twilight of the Establishment Clause

Ira C. Lupu F. Elwood and Eleanor Davis Professor Emeritus of Law, George Washington University Law School
Robert W. Tuttle David R. and Sherry Kirschner Berz Research Professor of Law and Religion, George Washington Law School

Americas Flag, Old Glory Flies Freely In A Country Churchyard in Hope New Jersey

[A version of this post will also appear at On the Docket, an online publication of the George Washington Law Review]

In Carson v. Makin, the Supreme Court held that Maine violated the Free Exercise Clause by excluding certain religious schools from a program that allowed parents to direct state funds to non-public schools. The case involved a challenge to a program that permitted parents in some rural school districts, which lacked adequate public schools, to use state subsidies for certain private schools. The statute did not categorically bar religious schools from eligibility to receive these funds, but it required participating schools to have a “nonsectarian” curriculum. The Supreme Court, with all six conservative justices joining the majority opinion, held that the exclusion of schools with a sectarian curriculum violates the Free Exercise Clause by subjecting those religious schools to unequal treatment without adequate justification.

Chief Justice John Roberts is not typically the author of opinions filled with smug disdain. But his opinion for the Court in Carson v. Makin certainly has that tone. The Court’s opinion treats the case as nothing more than a self-evident application of its prior rulings in Trinity Lutheran v. Comer and Espinoza v. Montana Department of Revenue. Both decisions rejected state laws, which had been based on state constitutional provisions, that excluded religious institutions from certain programs of governmental funding. Unlike Carson, however, those decisions focused specifically on the challenged statutes’ exclusion of religious entities because of their status – i.e., their religious identity. The opinions suggested, at least implicitly, that restrictions based on religious use of government funds would pose a different problem. In Carson, the majority opinion brushed aside that distinction with no explanation. “[T]hose decisions never suggested,” Chief Justice Roberts wrote, “that use-based discrimination is any less offensive to the Free Exercise Clause.”

However true that may be, use-based “discrimination” lies at the heart of traditional Establishment Clause limits on state funding of religious activity. The government may not sponsor or fund activities that are specifically religious – worship, religious instruction, and proselytizing. Or so constitutional lawyers thought for the past 75 years. The Court’s decision in Carson v. Makin has potentially opened a new world of mandatory government funding of specifically religious activity, throwing off many decades of precedent that tried to mesh equal treatment of religious institutions with the Establishment Clause’s restriction on government support for religion. Whatever balance might have existed before Carson is now history. The Court has decisively subordinated any Establishment Clause concerns to the right of equal funding under the Free Exercise Clause.

Carson v. Makin – The Decision

Those who follow the Court’s Religion Clause jurisprudence should not have been surprised at the decision in Carson v. Makin. The case involved a program of indirect aid – voucher financing – for non-public schools. The program required eligible schools to offer only “nonsectarian instruction.” That statutory limit dated back to 1981. Before then, parents could direct state tuition funds to any accredited school. In 1981, however, Maine’s legislature voted to exclude sectarian schools out of concern that such funding would violate the Establishment Clause. The concern was reasonable at the time, but the Court’s 2001 decision in Zelman v. Simmons-Harris resolved any doubts about the constitutionality of including accredited religious schools in a voucher financing scheme.

After describing the Maine program, the Court concluded that “[t]he unremarkable principles applied in Trinity Lutheran and Espinoza suffice to resolve this case.” Those unremarkable principles included (1) a determination that religious and secular private institutions are fully entitled to equal treatment under the Free Exercise Clause; and (2) the peremptory judgment that states do not have reasonable, much less compelling, independent grounds for imposing more restrictive non-establishment rules than those required under federal constitutional law. The second “principle” produces the Court’s disdainful treatment of Maine’s argument that the Court should respect its independent concerns about church-state separation, but it offers nothing new. The same arguments from Missouri and Montana met an identical fate in Trinity Lutheran and Espinoza. No respect for federalism is to be found here.

The first “principle” deserves a closer look. It is true that Trinity Lutheran and Espinoza rest on the Free Exercise Clause’s ban on discrimination against religion. But the cases seemed to recognize a limit on that principle, one that could only originate in the Establishment Clause. Categorical discrimination against a person or entity because of religious identity presumptively violates the Free Exercise Clause. Restrictions on government funding of specifically religious uses, however, such as worship or religious instruction, cannot be subject to the same categorical presumption of unconstitutionality because Establishment Clause jurisprudence had long imposed its own distinction between religious and secular uses of government funding.

The Court could have resolved this case with a narrower opinion that focused on the practical irrelevance of the status/use distinction for the Maine school funding program. The challenged program involves entirely indirect aid; parents are free to choose any (nonsectarian) school. The included private schools are free from nearly all controls over curriculum, state evaluation, and teacher qualifications. In other words, the state has no plausible responsibility beyond the minima of accreditation for the content of instruction that students receive in these schools. Thus, by focusing on the indirect character of the aid, the Court could plausibly have said that the link between the state and any potential religious use is broken by the private choice of parents. In that respect, Carson and Espinoza really are indistinguishable, and closely akin to Zelman.

Chief Justice Roberts did not choose this minimalist route for the decision, despite the majority opinion’s relative brevity and lack of engagement with counterarguments. Instead, he confirmed that the Court has firmly embraced the supremacy of free exercise interests over concerns about non-establishment of religion. The opinion thus stands in a long line of those, dating back to Justice Stanley Reed’s dissent in Illinois ex rel. McCollum v. Board of Education (1948), that have derided non-establishment norms as discrimination against religion in the public square.

The most important signal of this embrace of a long-rejected constitutional approach comes in the Court’s newfound refusal to recognize that distinctions based on religious identity differ profoundly from distinctions based on religious use. If the government must fund specifically religious activities when it funds analogous secular activity, what is left of the Establishment Clause in the funding context?

The New World of Religion Clause Law

At the time we write this post, the Court has under advisement Kennedy v. Bremerton School District, so for the moment we focus only on implications of this New World for government funding of religion. Kennedy, in which a public high school football coach claims a free exercise right to pray on the fifty-yard line immediately after games, may impose the emerging doctrines of the New World on speech by government agents in public schools. Funding alone, though, raises more than enough worries for those who retain some idea of a government limited in its capacity to promote religion.

In Carson, the Court suggests that Maine is not required to fund religious education. The majority offers a range of alternatives, primarily focused on expansion of the public education system either by bringing students to school (boarding schools) or bringing school to students virtually or in-person. Notably, however, the Court does not bless an alternative funding method Maine now permits in rural school districts that do not have a secondary school. Instead of allowing parents to choose the child’s school, the second method authorizes local education authorities to contract with a public or private school and to pay the full tuition of children who attend that “contract school.”

Following Carson, would Maine violate the Free Exercise Clause if it required private contract schools to conform to restrictions on religious teaching and activity that now apply to public schools? Contract schools involve direct aid, not parental choice – except the parent’s choice to place the child in a school without public funding – so does that take the practice outside the reach of Carson? Perhaps. But a contract solicitation that conditions eligibility on the school’s commitment not to engage in religious activity now seems the same as exclusion based on the school’s religious identity. Indeed, Trinity Lutheran involved a program of direct funding, and the Carson majority certainly did not discuss any relevant differences between the two cases. Despite the Court’s repeated emphasis on “parental choice” in Carson, we have no confidence that the distinction between direct and indirect aid has any more vitality after this decision than the distinction between religious status and use.

Although attention to the Court’s decision in Carson has focused almost entirely on its potential implications for education, the decision covers another broad domain of public spending – social welfare programs operated by private entities, including many faith-based organizations. Since the Charitable Choice legislation of the 1990s and President George W. Bush’s Faith-Based Initiative, the federal government has emphasized the right of religious institutions to participate in a wide variety of social welfare funding programs. Such programs include substance abuse treatment, adoption and foster care, social services that support job training and parenting skills, among many others. The rules that have governed these programs for the past two decades guard against discrimination based on religious status. Religious entities have the right to compete on equal terms for government funds. In light of longstanding constitutional norms, however, the rules explicitly forbid funding for religious uses of the funds. Providers may use government funds only for programs that do not include “specifically religious activities,” typically delineated in statute and regulation as prayer, worship, or religious instruction. Any provider that also engages in such activities must segregate them in time or space from the government-funded program, and delivery of program benefits may not be conditioned on the beneficiary’s willingness to participate in those activities.

Until the decision in Carson, we had good reason to believe that these conditions on government funding were required by the Establishment Clause. Now, however, those conditions have been called into serious question. A religious provider that believes it cannot confine its social welfare programs to exclusively secular components has always been categorically ineligible to apply for direct funding. After Carson, those providers will certainly challenge the program rules under the Free Exercise Clause. If conditions imposed on religious use of government funds are the same as discrimination based on religious identity, then government must eliminate those restrictions on aid for faith-intensive social welfare programs.

In eliminating such restrictions, government will inevitably confront problems with Free Exercise claims by program beneficiaries who want to receive government benefits without the incorporated religious messages. Governments at all levels will be forced into the impossible choice of trying to accommodate those individual objections, out of concern about the welfare of program beneficiaries who could be steered into unwanted religious experience, or moving the programs, long based on public-private partnerships, entirely into public institutions. As Justice Sotomayor wrote in dissent in Carson, this inevitably will result in a decrease in total services available.

Moreover, the Court’s decision in Carson calls into question other conditions on government funding. As we have seen many times before, including in last term’s decision in Fulton v. City of Philadelphia, government may include contract provisions – such as prohibitions on discrimination based on sexual orientation or gender identity – that some religious entities find objectionable. The Court resolved Fulton on seemingly narrow grounds, but Carson may give that decision a much firmer foundation in the Free Exercise Clause. A condition on receipt of government funds that effectively excludes one or more religious providers would once again invite strict scrutiny. That is what Justice Alito argued for in Fulton, though he assumed that the Court would need to overturn Employment Division v. Smith to reach that conclusion. By reshaping the non-discrimination principle into a broad primacy of free exercise over non-establishment and other legitimate governmental interests, Carson may undo such conditions without overturning Smith. Where there is a judicial will, there is a judicial way, and the Court’s treatment of free exercise claims in recent years suggests an indomitable will to prefer and protect the interests of religious actors over the competing interests of others.

The only solace we can offer to those who care about sound constitutional principles of religion-state relationships is this. Over time, any constitutional understanding that promises equal benefits to religion without imposing equal obligations on religion is likely to collapse from lopsided favoritism. The Religion Clauses demand symmetry – the government should not finance what it cannot regulate, and vice versa. This is why we defend the ministerial exception, which bars government regulation of the employment relationship with those who preach a faith. Government should neither regulate nor subsidize exclusively ecclesiastical activities, such as worship. A Court that demands public support to religion without holding it to account to public values will eventually lose its legitimacy on this subject, as the Roberts Court is at risk of doing on other subjects as well.


Dissenting opinions rarely understate the significance of the decision to which they object. But the dissents of Justices Breyer and Sotomayor do just that. Justice Breyer expresses concern that the majority has eliminated the “play between the joints” of the Religion Clauses, but that metaphor had been relegated to the facts of Locke v. Davey before Carson. Justice Sotomayor objects that the Court is dismantling the “wall of separation” between church and state, but that metaphor lost its salience long before Carson. The dissenters have missed or understated the radical, revolutionary character of the Court’s decision. With dismay, we have been observing this precise trajectory in Religion Clause law, but we are surprised that the New World arrived so soon, and with so little effort to explain the supposed defects of the regime it replaces.

Establishment Clause (or Separation of Church and State)